>>11
It's not just Chrome, all the major web browsers have implemented their own subset of the standard that isn't in compliance with the whole spec or each other. The issue of browser specific content like activex and all of the "Best viewed from Netscape 3+" pages from the 90's never was unified, it just fails silently now or there's a huge mess of javascript errors because just using more of that cancer of a language is apparently a patch for the standards being bad and poorly implemented. The relatively good standards that are out there excel because they are simple ubiquitous, you pipe from a to b and you can see how and why it's doing that and it's not checking for and doing 30 different things to finish the job, "comprehensive" is a bad word and all that cleverness that led to HTTP/2.0 or even 1.1 seems to have been wasted adding more specific cases than cutting the fat and letting those be implemented with broader strokes. The original Xanadu Hypertext design document was overcomplicated but at least the end user's view of it was something impressive, the current W3C hypertext doc is voynichian and what it accomplishes is mundane, and the IETF's transfer protocol for it accomplishes little for great undue complexity.